Back Matter

Back Matter

Robert Sharer, and Piritta Sorsa
Published Date:
February 1998
  • ShareShare
Show Summary Details
Appendix I Methodology

This appendix presents the methodological basis for the review of trade policy in Fund-supported programs, including the selection of Fund arrangements for review, the 10-point scale for classifying the restrictiveness of trade regimes, and the selection of countries for detailed case studies. The approach is also compared with other studies on trade reform in developing countries.

Overview of the Basic Approach

A large number of programs supported by Fund arrangements are reviewed to provide a basis for assessing the ambitiousness of trade reform in Fund-supported programs and the implementation record. The trade regime of each country is examined at the outset of the Fund-supported program and its overall trade restrictiveness is assessed, based on a 10-point scale.

The 10-point classification scheme combines measures of the trade restrictiveness of import tariffs and NTBs. Five ranges are specified for import tariffs, with the lowest range (0–10 percent) being the least restrictive and the highest range (25 percent or over) being the most restrictive. Three categories have been specified for NTBs, ranging from open, to moderate, to restrictive. These categories are based mainly on the number of sectors covered by NTBs (whether NTBs are confined to a few sectors of the economy), on the production (whether they cover entire stages of production) or trade coverage of these barriers, and on their restrictiveness.

The five classifications of import tariffs and three classifications of NTBs yield a 15-cell matrix. Within the matrix the cells were converted to a 10-point scale by assigning a relative ranking, from 1 to 10, representing the overall restrictiveness of the trade regime. The most open import tariff and NTB regimes were assigned the number 1, and the most restrictive tariff and NTB regimes were assigned the number 10.

The ambitiousness of trade reform can be assessed by identifying the trade content of each Fund-supported program and assessing how far along the 10-point scale these measures would have moved the trade regime if fully implemented. The implementation of trade reform is assessed by evaluating the overall restrictiveness of the regime at the end of the program period, again based on the ten-point scale.

A number of other questions are covered in the review of each arrangement. Were specific medium-term targets for trade reform included in the original program? How were different types of monitoring instruments used and how did they influence implementation of trade reform? What was the role of the World Bank in respect of trade reform for the programs covered? What role did the GATT/WTO and other international institutions play? Did membership in an RTA promote or impede liberalization? How did fiscal and balance of payments considerations influence trade reform? A number of these questions are taken up in more detail in the case studies.

Selection of Fund Arrangements for Review

The review covers all Fund arrangements of two years or more in length that started after January 1, 1990 and that ended by June 30, 1996. Arrangements of less than two years duration are excluded since these typically focus on financial stabilization and would be expected to have less emphasis on structural policies, including trade policy reforms. The cutoff point of mid-1996 was selected to ensure sufficient time had elapsed since the conclusion of the arrangement to permit a proper assessment of performance under the Fund-supported program. On this basis, 30 arrangements supporting 28 programs for 27 countries are reviewed. These include arrangements under the EFF, SAF, and ESAF,57 as well as two RAPs, and four Stand-By Arrangements.

Country Case Studies

Six countries were selected for detailed case studies to complement the review of trade reform in the context of 28 Fund-supported programs. The case studies cover the 1990–96 period, with earlier periods covered if they relate directly to trade reform in 1990–96, and consider the following questions. In cases where the Fund-supported programs were not very ambitious, did Fund staff initially propose a more ambitious trade reform in the program negotiations? If trade reform went off track, what factors may have been responsible? In particular, did fiscal constraints, balance of payments problems, or obligations to regional groupings, affect trade reform? What characteristics did successful trade reformers possess that were absent in less successful reformers (e.g., a well-developed tax system)?

Based on results of the reviews of 28 Fund-supported programs, countries were classified into four broad categories: (1) liberalizers where the program included reasonable trade policy conditionality; (2) liberalizers where the program did not include reasonable trade policy conditionality; (3) nonliberalizers where the program included reasonable trade policy conditionality; and (4) nonliberalizers where the program did not include reasonable trade policy conditionality.

Countries with good information on the nature of their trade regime were selected with a view to covering three of the four broad categories (category 2 is of little relevance for the evaluation of Fund programs). The final selection of countries (Bangladesh, Egypt, Hungary, Sri Lanka, Zambia, and Zimbabwe) also took account of a second set of desirable subcriteria:

  • nature of conditionality for monitoring trade reform implementation in the program;

  • membership in regional trading arrangements that may have influenced the pace of trade reform;

  • fiscal or balance of payments constraints; and

  • inclusion of countries from different regions of the world.

Overall Trade Restrictiveness Classification Scheme

Table 12 illustrates the assignment of tariff and NTB categories on the 10-point scale. The assignment places more weight on restrictiveness of NTBs, which are inherently less transparent and more distortionary than tariffs.

Table 12.Classification Scheme for Overall Trade Restrictiveness
Nontariff Barriers
Relatively open258
Relatively restrictive4710
Source: IMF staff estimates.
Source: IMF staff estimates.

The restrictiveness of the import tariff regime depends on many factors, including the minimum and maximum tariff rates, the number of bands, the allocation of individual items to the bands, existence of “exceptional” rates that lie outside the basic tariff structure, any other duties and charges (such as differential rates of excise or VAT taxes on imports, import surcharges, and statistical fees), and the extent of customs duty exemptions. The amount of information available on the tariff regime varies considerably among the countries. In many, there was enough information to compute an average tariff rate but insufficient information to assess the degree of tariff dispersion or the prevalence of discretionary customs duty exemptions (as opposed to statutory zero customs duty rates). In view of these limitations, the study adopted a pragmatic approach and used the average import tariff rate as a basis for classifying the tariff regime.58

The five-category ranking of average import tariffs, based on the average tariff rates of Fund member countries, is specified in Table 13.

Table 13.Classification Scheme for Tariff Restrictiveness
RestrictivenessAverage Tariff Range (t)
Open10≤t<10 percent
Relatively open10≤t< 15 percent
Moderate15 ≤t<20 percent
Relatively restrictive20 ≤t<25 percent
Restrictive25 percent or over
Source: IMF staff estimates.
Source: IMF staff estimates.

Broadly equal numbers of Fund members fell into each of these ranges based on the latest available information on average import tariff rates. Alternative ranges were considered in connection with the results of the review.

Countries employ a wide variety of restrictions on imports and exports other than tariffs, including import/export quotas, restrictive licensing, bans, state trading/monopolies, restrictive foreign exchange allocation, and multiple exchange rates. Such measures in effect provide indirect subsidies to import-competing domestic producers in a nontransparent manner. Information on the presence of NTBs and their restrictiveness (as measured for instance by ad valorem equivalents)59 may be limited;60 the review of previous studies on trade reform in developing countries below shows that others have faced similar difficulties. In view of this, Table 14 shows the three ranges of NTBs61 that have been used in this study.

Table 14.Classification Scheme for NTB Restrictiveness
NTB RegimeClassification Criteria
OpenNTBs are either absent or minor. Less than 1 percent of production or trade is subject to NTBs.
ModerateNTBs are significant, covering at least one important sector of the economy (e.g., agriculture or textiles) but not pervasive (e.g., all consumer goods).
Between 1 percent and 25 percent of production or trade is subject to NTBs.
RestrictiveMany sectors or entire stages of production (e.g., all consumer goods) are covered by NTBs.
More than 25 percent of production or trade is subject to NTBs.
Source: IMF staff estimates.
Source: IMF staff estimates.

While more trade variables (such as tariff dispersion or customs duty exemptions) could in principle be included in the overall index of trade restrictiveness, by adding dimensions to the 15-cell matrix, data difficulties and added complexity strongly support the approach of omitting these factors from the overall index. In principle, it would also be preferable to assess the effects of trade restrictions on prices and resource allocation, but this is very difficult and time-consuming even for industrial countries; it is not feasible for the program countries analyzed in this review. The review of previous studies below illustrates these points.

Comparison with Approaches in Previous Studies

A number of previous studies have attempted to quantify the extent of trade reform in developing countries. Some have not attempted to form an overall index of trade restrictiveness (Dean, Desai, and Riedel, 1994); others have developed simple indicators of openness (Kirmani and others, 1994b; Papageorgiou, Michaely, and Choksi, 1991; and Sachs and Warner, 1995); while others have formed quantitative measures that summarize the overall stance of the trade and exchange regime based on multiyear research projects (Balassa, 1971; Bhagwati, 1978; and Krueger, 1978). This section examines these previous efforts in relation to the methodology used in the present study to assess the ambitiousness and implementation of trade reform in Fund-supported programs.


Bela Balassa (1971) led a research project on the structure of protection in six developing countries (Brazil, Chile, Mexico, West Malaysia, Pakistan, and the Philippines) and, for comparison, one developed country (Norway). The study analyzes the pattern of trade protection in these countries and its implications for resource allocation, exports, and economic growth.

Measures of trade protection were developed, known as effective rates of protection, that aimed to assess how trade protection affected production incentives for different industrial activities:

  • Effective rates of protection took account of both nominal rates of protection for imported intermediate inputs and the nominal protection afforded to the final products.

  • If import QRs were binding rather than import tariffs, then the differences between domestic and world prices were often used, as measures of the price equivalents of the QRs, to compute effective rates of protection.

Notwithstanding the well-known theoretical limitations of effective protection measures as indicators of resource misallocation, as noted by Bhagwati and Srinivasan (1973), the study makes clear that the calculation of effective rates protection measures require very substantial amounts of information and analysis.

Bhagwati and Krueger

Jagdish Bhagwati (1978) and Anne Krueger (1978) led a study of liberalization of trade and exchange controls. Extensive case studies were prepared for each of the following ten countries: Brazil, Chile, Colombia, Egypt, Ghana, India, Israel, the Philippines, South Korea, and Turkey.

The study analyzes the balance of payments effects of exchange rate devaluation for developing countries with complex exchange control regimes (including trade and exchange restrictions). It decomposes the benefits of moving to a more liberal trade and exchange system following a nominal devaluation into three parts:

  • avoidance of static costs of QR regimes due to inefficiencies in licensing and resource misallocation;

  • higher growth stemming from increased exports under a more realistic exchange rate; and

  • higher growth stemming from improved resource allocation due to reduced antiexport bias of the trade and exchange regime.

To quantify these benefits, detailed information was required on the nature of the trade and exchange regime. For instance, effective exchange rates for exports and imports were calculated to evaluate the extent of antiexport bias; these involved adjusting exchange rates for import tariffs, import surcharges, differences in domestic and foreign prices, and export subsidies.

The study led by Bhagwati and Krueger illustrates the extent of information and analysis required to calculate measures that summarize the effects of the trade and exchange regime on resource allocation and economic growth. The present study focuses on trade liberalization and generally excludes consideration of the exchange regime, except where exchange restrictions clearly act as binding trade restrictions.

Papageorgiou, Michaely, and Choksi

Papageorgiou, Michaely, and Choksi (1991) analyze trade liberalization in 19 countries during 36 trade reform episodes. They consider a wide range of issues, including the extent of trade liberalization as measured by movement along an index that evaluated intervention in the trade regime. The index ranges from 1 to 20, with 1 being the highest possible degree of intervention and 20 being the most “neutral”:

  • A completely neutral trade regime would prevail if there were no government interference in the trade regime. A value was assigned for each year to track progress toward neutrality over time;

  • Authors developed the indices for individual countries based on sets of quantitative or qualitative information that were judged relevant to moving the economy closer toward full neutrality;62

  • It was not possible to use a common methodology and therefore the indices are not comparable across countries;

  • The indices are not intended to provide quantitative measurements.

The present study, in contrast, uses a common methodology for constructing indices of overall trade restrictiveness in each country.

Dean, Desai, and Riedel

Dean, Desai, and Riedel (1994) provide a recent quantitative study of trade liberalization in developing countries. It examines, inter alia, changes in tariffs and NTBs between 1985 and 1993 for the following 32 developing countries: Argentina, Bangladesh, Brazil, Cameroon, Chile, China, Colombia, Costa Rica, Côte d’Ivoire, Democratic Republic of Congo, Ghana, India, Indonesia, Kenya, Korea, Madagascar, Malawi, Malaysia, Mali, Mexico, Nigeria, Pakistan, Peru, Philippines, Senegal, South Africa, Sri Lanka, Tanzania, Thailand, Uganda, Venezuela, and Vietnam.

Where available, data are presented for each country on tariff levels, range, and dispersion, and coverage of quantitative trade restrictions. These measures are used separately to assess progress in trade liberalization, including whether there has simply been a shift in the type of trade protection (e.g., tariffs were used to replace QRs) or whether the level of protection has actually declined. There is, however, no attempt to form an overall measure of trade restrictiveness.

Kirmani and Others

The method of classifying the overall restrictiveness of the trade regime in the present study is similar but more detailed than that employed in the study of trade reforms in Fund-supported programs by Kirmani and others (1994b):

  • Tariff regimes are classified based on the average statutory tariff (including other duties and charges), with rates of 10 percent or below classified as relatively open, 10–25 percent as moderate, and above 25 percent as restrictive. The earlier study did not incorporate tariff dispersion or exemptions into the classification of trade regimes due to measurement difficulties;

  • The restrictiveness of import/export QRs is also classified into three categories based on the proportion of tariff lines subject to QRs as well as the intensity of the QRs, in a manner broadly similar to that used in the present study.

The separate classifications of the tariff and QR regimes are combined into the matrix shown in Table 15:

Table 15.Overall Classification Scheme Used in the Study by Kirmani and Others
Quantitative Restrictions
Relatively openModerateRestrictive
Relatively openOpenModerateRestrictive

The classification into only three categories of overall trade restrictiveness provides limited scope to distinguish between the initial position of countries’ trade regimes, or to capture movements as trade reform measures are implemented. The more detailed ten-point classification used here provides greater scope to capture changes in overall trade restrictiveness due to trade reform implementation.

Another important difference is that the 1994 study compared, inter alia, the overall restrictiveness of the trade regime for program countries between the end of 1990 and the end of 1993. It did not link any changes in overall restrictiveness to the policy content of the Fund-supported program or its implementation. The present study evaluates the restrictiveness of trade regimes at the outset of the program period for each arrangement, assesses how far the specific trade reform measures included in the Fund-supported program would have moved the trade regime along the ten-point scale if they had been fully implemented, and evaluates restrictiveness of the trade regime at the end of the program period. This methodology examines the trade content of Fund programs, how ambitious they were, and their actual implementation.

Sachs and Warner

Sachs and Warner (1995) consider the relationship between trade openness and economic performance, including notably economic growth. Of concern in the present context is the measure of openness used in the study. A country is judged to be closed if it has at least one of the following characteristics:

  • NTBs cover at least 40 percent of trade;

  • average import tariffs are at least 40 percent;

  • the black market exchange rate is depreciated by at least 20 percent relative to the official exchange rate, on average, during the 1970s or 1980s;

  • a socialist economic system (as defined by Kornai (1992)); or

  • a state monopoly on major exports.63

The study defines an open economy as one in which none of the above five conditions apply. Data on tariffs and NTBs for 1985–88 (taken from Barro and Lee, 1994) who rely mostly on data from the United Nations Committee on Trade and Development (UNC-TAD) were used to construct the measures of openness, along with an independent literature review to judge the timing of shifts from a closed to an open trade regime.

Sachs and Warner acknowledge that the chosen criteria would benefit from further refinement, for the following reasons:

  • the threshold levels are arbitrary (e.g., a black market premium of at least 20 percent);

  • nominal rather than effective rates of protection are used;

  • the effect of export subsidies in offsetting the anti-export bias of import protection is not considered; and

  • quantification of the effects of NTBs is inherently difficult (the study relies mostly on measures of NTBs used by UNCTAD).

Comparison of the Index with Other Indices

Based on this review of the main studies of trade liberalization in developing countries, the following points can be made regarding the usefulness of the index of overall trade restrictiveness used here in comparison with other indices developed in previous studies. First, the index is more detailed than the binary openness indicator developed in Sachs and Warner (1995). It provides ten overall classifications, in contrast to the three overall ratings distinguished in Kirmani and others (1994b), as well as separate ratings for tariffs and NTBs. Second, the index provides a synthesis of tariff and NTB data that is useful for making policy judgments, compared with that used by Dean, Desai, and Riedel (1994), which tabulates tariff and NTB measures but provides no overall ratings. Third, the index provides a reasonable basis for cross-country comparisons owing to the considerable efforts to develop a common analytical framework, in contrast to the 20-point scales of Papageorgiou, Michaely, and Choksi (1991) that were decidedly not comparable across countries. Fourth, although the index is perhaps not as detailed and precise as the effective rates of protection in Balassa (1971) or the effective exchange rates in Bhagwati (1978) and Krueger (1978), construction of such measures is a massive undertaking for each country case considered. This comparison suggests that the index used here is preferable to those feasible alternatives developed in previous studies.

Appendix II Summary of the Detailed Case Studies

The findings of the six country case studies that examine trade reform in Bangladesh, Egypt, Hungary, Sri Lanka, Zambia, and Zimbabwe during 1990–96 are summarized below.


During the period 1990–96, Bangladesh had one Fund arrangement, followed by a program monitored under enhanced surveillance. The ESAF arrangement covered in the review was approved in August 1990 and covered the period through June 1993, and enhanced surveillance was in place during 1993–94.

Notwithstanding a SAF arrangement in 1986/87–1988/89, Bangladesh’s trade regime was highly restrictive at the outset of the ESAF arrangement. The unweighted average tariff was estimated at around 89 percent, the maximum rate was over 500 percent, and there were 24 tariff bands. In addition, a development surcharge of 8 percent applied to all imports. Import NTBs were extensive, with 315 tariff line items (or about 25 percent of the total) subject to some form of quantitative restriction. Export NTBs included export restrictions or outright bans on about 20 items, including rice.

Significant progress was achieved in reducing the restrictiveness of Bangladesh’s trade system during the ESAF arrangement period, yet it remained restrictive. The unweighted average tariff declined to an estimated 59 percent, the maximum tariff rate fell to 100 percent, the number of tariff bands was reduced to 15, and the development surcharge was eliminated. Also, the number of tariff line items subject to import NTBs was reduced to 93 (or about 7 percent of the total). Jute export subsidies and the minimum export price for jute products were eliminated, but raw jute exports were still subject to an export price check and, more important, export restrictions or outright bans remained. Thus, the overall restrictiveness of Bangladesh’s trade regime fell from a rating of 10 in 1990 to 8 in 1993 (Table 16).

Table 16.Bangladesh: Restrictiveness of Trade Regime
Programn. a.1088n.a.n.a.n.a.
Source: IMF staff estimates.Scale: 1 (least restrictive) to 10 (most restrictive). n.a. denotes not applicable.
Source: IMF staff estimates.Scale: 1 (least restrictive) to 10 (most restrictive). n.a. denotes not applicable.

Subsequently, considerable further progress was achieved in tariff reduction in the context of the enhanced surveillance program in 1993/94 and of the World Bank Industrial Sector Adjustment Credit (ISAC II). The maximum tariff rate was reduced to 45 percent by 1996 and customs duties on various capital goods, raw materials, and inputs for agriculture and selected industries were either eliminated or reduced. As a result, the unweighted average tariff declined to about 22 percent by 1996/97 and the number of tariff bands fell to 7. However, the number of tariff line items subject to import NTBs rose to 110 (or about 9 percent of the total). Export restrictions or outright bans remained. As a result, the overall restrictiveness of Bangladesh’s trade system declined to a rating of 7 by 1996.

Targets and Implementation Under the Fund Program and Trade Reform in Nonprogram Years

The ESAF arrangement sought to rationalize the tariff structure by reducing the maximum rate (except for some luxury goods) and the number of bands, and to phase out import NTBs. In addition, bans and restrictions on imports were to be reduced substantially, except for a small number of “sensitive” items. Virtually all import restrictions on items required by exporters were to be removed and the existing duty drawback system for exports was to be strengthened. The elimination of export subsidies was envisaged, except for jute exports; the latter were to be phased out on a more gradual basis. Trade reform targets, most of which were monitored via prior actions, were met.

Subsequently, there was some further liberalization of Bangladesh’s trade regime in the context of the enhanced surveillance program and the World Bank ISAC II. The latter provided for, inter alia, removal of import controls, rationalization of the tariff structure, implementation of trade neutral taxes, including the VAT, and a strengthening of customs administration. As noted above, tariffs were reduced, but import NTBs rose.

Factors Affecting Design and Implementation of Trade Reform in Fund Program

A major factor affecting the pace of trade reform was the authorities’ concern about the negative impact of trade liberalization on the balance of payments and fiscal positions (customs duties accounted for more than 30 percent of tax revenue). Equally important, the negotiations on trade reform under the ISAC II with the World Bank were protracted and affected the quality of medium-term trade reform targets in the ESAF program, which were to be based on the outcome of these negotiations. Also, the importance of the textile industry hampered removal of NTBs protecting this sector. Prior actions were used to monitor the implementation of trade reform under the ESAF arrangement to highlight the importance of trade reform.

In general, other factors such as membership in RTAs and commitments to the GATT/WTO appear not to have affected the pace of trade reform. Bangladesh’s tariff bindings under the Uruguay Round were well above the applied rates and would not have entailed further liberalization of its trade regime. The overall macroeconomic framework remained on track throughout, with all financial and structural performance criteria observed on a timely basis.


While considerable trade liberalization took place, especially on the import tariff structure, Bangladesh’s trade system remained relatively restrictive at the end of 1996. The critical nature of trade reform was reflected in the use of prior actions in the ESAF arrangement and most trade reform targets were met in a timely manner. A key factor hindering trade reform, however, was the authorities’ concern about the potentially negative effect of trade reform on the fiscal and balance of payments positions. Protracted negotiations between the authorities and Bank staff resulted in a weak and less focused specification of medium-term trade reform targets under the ESAF arrangement.

Bangladesh’s experience points to the need to implement trade reform in the context of a comprehensive macroeconomic adjustment program to, inter alia, reduce pressures on the fiscal and external positions. In this regard, the lack of well-specified, medium-term trade reform targets and supporting intermediate steps may have impeded trade reform. Given Bangladesh’s heavy reliance on trade taxes, early adoption of a trade-neutral tax structure, including the VAT, might have helped trade reform to proceed without unduly affecting the fiscal position. It also points to the ability of an important industrial sector to resist early dismantling of import barriers that protect it.


During the period 1990–96, Egypt had three Fund arrangements. Besides the May 1991 Stand-By Arrangement covered in the review, which covered the period through June 1993, a successor three-year EFF arrangement was approved in September 1993. A two-year Stand-By Arrangement was approved in October 1996.

Egypt had a restrictive trade system in 1990, ranked 10 on the scale of trade restrictiveness (Table 17). The general tariff schedule ranged from zero to 120 percent, but there were exceptions with substantially higher rates applied to a few “luxury” and other goods that included cosmetics, automobiles, alcohol, and tobacco. The unweighted average tariff was estimated at about 35 percent in 1990. Although Egypt’s import licensing scheme was abolished in 1986, import bans covered about one-third of tradable goods output in late 1990. Also, some 56 tariff line items required import prior approvals. There were export taxes on nine tariff line items, but some of these were not binding as the same products were covered by bans. In late 1990, export prior approvals and export bans covered, respectively, 8 percent and 7 percent of tradable goods output. In addition, a further 17 tariff line items—covering about 1 percent of tradable goods output—were subject to export quotas.

Table 17.Egypt: Restrictiveness of Trade Regime
Source: IMF staff estimates.Scale: 1 (least restrictive) to 10 (most restrictive). n.a. denotes not applicable.

Under the Stand-By Arrangement of 1996–98, the restrictiveness of Egypt’s trade regime is expected to decline to a rating of 4 by mid-1998.

This rating refers to the target under the 1993 Extended Fund Facility (EFF) for the end of 1995.

This rating refers to the target under the 1996 Stand-By Arrangement for the end of 1996.

Source: IMF staff estimates.Scale: 1 (least restrictive) to 10 (most restrictive). n.a. denotes not applicable.

Under the Stand-By Arrangement of 1996–98, the restrictiveness of Egypt’s trade regime is expected to decline to a rating of 4 by mid-1998.

This rating refers to the target under the 1993 Extended Fund Facility (EFF) for the end of 1995.

This rating refers to the target under the 1996 Stand-By Arrangement for the end of 1996.

Despite progress with trade liberalization in 1990–96, Egypt’s trade system remained restrictive at the end of the period and ranked 8 on the scale of trade restrictiveness at the end of 1996.64 The general tariff schedule ranged from zero to 55 percent. Substantially higher tariff rates applied to alcohol, tobacco, and automobiles, and import surcharges of 2 percent or 4 percent applied to all imports. The unweighted average tariff was estimated at above 25 percent. While import bans were substantially reduced during the Stand-By Arrangement of 1991–93,65 there is evidence that some have been replaced by restrictive quality controls. In addition, there are cumbersome customs procedures. All export taxes were eliminated in late 1992, and the only remaining export NTB is a ban on exports of raw hides, which is scheduled to be removed in 1998.

Targets and Implementation Under Fund Programs

The program for the 1991 Stand-By Arrangement aimed to reduce the trade regime’s restrictiveness by 2 points. The range of general tariff rates was to be reduced to 10–80 percent by June 1992, and thereafter tariff reform was expected to reduce further dispersion and the average rate, as well as to eliminate exceptions to the general tariff structure and to reduce preferences. On import NTBs, import bans were scheduled to be reduced significantly by June 1992, and import prior approvals were to be phased out. Export bans would be reduced, export quotas increased, and export prior approvals eliminated.

Most trade reform measures were implemented as envisaged in the early part of the program. Restoration of customs duty rates to their early-1989 levels (within the new general tariff schedule) was a prior action and an integral part of the fiscal revenue package, and increased the average tariff to 42 percent. For the first review, trade reform was implemented as envisaged. The reduction and elimination of NTBs was greater than programmed. The second stage of trade reform, linked to the World Bank Structural Adjustment Loan (SAL), was delayed. The general maximum tariff was reduced with a delay, while the minimum rate remained unchanged.66 The program reduced tariff dispersion, but the unweighted average tariff remained high and was estimated at 34 percent at the end of the program. Substantial progress was made in removing both import and export NTBs, but the number of products subject to restrictive quality controls increased.

The 1993 EFF program aimed at substantial trade liberalization, that is, a movement of 4 points on the scale of trade restrictiveness. The maximum general tariff rate was to be reduced from 80 percent to 70 percent by December 1993 with a corresponding reduction in all tariff rates above 30 percent. Thereafter, tariff reform was expected to reduce dispersion and the average rate by lowering, in two stages, the maximum general tariff to 50 percent by December 1995 with corresponding reductions in tariff rates above 30 percent. Tariff rates at 30 percent or below were to remain largely unaffected, but tariffs on some capital goods were to be reduced to 5 or 10 percent by December 1995. All but one of the exceptions to the general tariff structure were to be eliminated by June 1994.

On import NTBs, all import bans were eliminated in June 1993, except on textiles, garments, and poultry. The import ban on poultry was to be eliminated by June 1994, and the other bans were to be addressed by the ongoing Uruguay Round negotiations. In addition, the authorities were committed to implement GATT rules for quality control standards by June 1994. Remaining export NTBs were to be eliminated by June 1995.

Reflecting the importance accorded to trade reform, the program relied heavily on prior actions to monitor trade policy implementation. They included the public announcement of Egypt’s medium-term trade reform plans. However, the structural reform component of the program was not implemented as envisaged; limited trade liberalization was accomplished; and the restrictiveness of Egypt’s trade system did not change substantially. Although the maximum tariff, tariffs on capital goods, and tariff dispersion were reduced, the unweighted average tariff remained high, estimated at 32 percent at the end of the program period. Also, import surcharges at the rates of 2 percent and 5 percent were introduced for fiscal revenue reasons in April 1994. No progress was made in eliminating the restrictive features of quality controls on imports, and, therefore, NTBs did not change significantly. Tariff reform was postponed on the grounds that it would have imposed an “unacceptable burden” on lower-income groups.

The 1996 Stand-By Arrangement program aims at a substantial reduction in the restrictiveness of Egypt’s trade system, amounting to a movement of 4 points on the scale mainly through the elimination of the remaining import NTBs and further tariff reductions. As in the EFF, the importance of the trade reform was signaled via prior actions. These prior actions were significant and included the public reaffirmation of the government’s commitment to medium-term trade liberalization and faster integration in the world economy, as well as significant tariff reductions in the general maximum rate (to 55 percent) and rates above 30 percent.67 Thereafter, the general maximum tariff rate is expected to be reduced in two stages to 40 percent by July 1998 (with tariffs between 30–50 percent to be reduced to 30 percent).68 In addition, the import surcharge will be reduced and unified at 1 percent by July 1998.69 Tariffs on automobiles were initially cut in October 1996 and are to be reduced over the next five years, but the other exceptions to the general tariff structure (alcohol and tobacco) are not scheduled to be eliminated during the program period. The import ban on poultry was eliminated in July 1997 and replaced with an 80 percent tariff. Egypt’s system of mandatory quality controls on imports is being reviewed and will be modified in the second half of 1997 to eliminate its restrictive features.

Factors Affecting Design and Implementation of Trade Reform in Fund Programs

The 1991 Stand-By Arrangement. Most of the trade liberalization measures in the program were designed under the auspices of the World Bank SAL. Fund staff noted on several occasions that a decision by the Bank to proceed with a SAL would be required to proceed with a Stand-By Arrangement, but there was little discussion of specific trade measures in the program, except the increase in custom duties for fiscal reasons. However, the balance of payments position was vulnerable and may have affected the design of trade liberalization measures. The second stage of trade reform was delayed because of employment and political economy considerations.

The 1993 EFF. The main trade-related measures in the program were designed in close collaboration with Bank staff and were part of the World Bank Structural Adjustment Monitoring Program. The negotiations over the structural reform measures in the EFF were protracted. The Fund staff favored ambitious trade reform, but the final program had a more gradual pace. In negotiations, the authorities were concerned that faster trade liberalization could worsen unemployment, although the need for liberalization in the longer term was recognized. Implementation was influenced by employment and political concerns.

The 1996 Stand-By Arrangement. Program negotiations were again protracted and centered on structural reforms. Although the main trade reform measures were designed based on earlier Fund-Bank work on trade liberalization, the Bank was not directly involved. In the negotiations, the Fund staff’s initial position in the area of trade reform was ambitious but the final program had a more gradual pace of trade liberalization. Throughout the negotiations, the authorities advocated a more measured pace of structural reforms, including trade reform, for reasons of political economy. They also expressed concern about the impact of reductions in tariffs and the import surcharge on the fiscal position.


Substantial trade liberalization was implemented by Egypt in the 1990s, with Fund and World Bank assistance, but its trade system remained restrictive at the end of 1996, although further liberalization is proceeding in 1997 and planned for 1998. Egypt’s macro-economic environment improved throughout most of the period. A more gradual approach to trade liberalization was adopted owing to fears that faster reform could have worsened unemployment in the short run and to maintain the domestic political consensus.

Egypt’s experience shows that trade reform can be implemented together with fiscal adjustment. It demonstrates the ability to reinject momentum in trade liberalization after a slowdown in these reforms. Also, it points to the need to adopt measures to address possible adverse effects of trade reform on domestic production and employment in the short run.


During the period 1990–96, Hungary had four Fund arrangements. The 12-month Stand-By Arrangement approved in March 1990 did not contain trade reform. It was replaced in February 1991 by the EFF covered in the review. A new Stand-By Arrangement was approved in September 1993 but did not contain trade reform; fiscal difficulties contributed to this program going off track in 1994. A two-year Stand-By Arrangement was approved in March 1996.

In 1990, Hungary had a restrictive trade system, ranked 10 on the scale of restrictiveness. The unweighted average tariff rate was about 21 percent (including 5 percent customs fees) with a high dispersion of rates. Tariff exemptions were widespread and, for example, about 40 percent of imports (from CMEA partners) were not subject to customs duties. About three-fourths of imports were subject to NTBs (mostly restrictive licenses), including a quota on consumer goods imports amounting to $200 million, or the equivalent to 2 percent of imports. Trading monopolies covered products accounting for about 20 percent of imports and 35 percent of exports. There were no export taxes, but all exports were subject to licencing. Export subsidies existed for CMEA trade and for agricultural exports. In addition, foreign exchange restrictions acted as a trade barrier.

Hungary’s trade regime was substantially liberalized during 1990–96 and ranked 6 on the scale of restrictiveness by the end of 1996 (Table 18). Most trade liberalization occurred in 1991 in the context of the EFF when the CMEA managed trade system moved to world prices, which resulted in a considerable reduction in NTBs. However, subsequent progress in reducing the remaining import NTBs, which cover 7–10 percent of imports (mostly linked to the consumer goods quota), has been limited. The remaining export licenses concern mostly products subject to restrictions in export markets (an estimated 30–40 percent of exports). The unweighted average tariff rate declined to 18 percent in 1991 (tariff 13 percent, customs fees 5 percent), where it remained until 1994. However, it increased to about 22 percent in 1995 when Hungary introduced an import surcharge of 8 percent, partly offset by a reduction in basic tariffs under its Uruguay Round commitments. In 1996, the unweighted average tariff rate was reduced to 18 percent as the surcharge rate declined to 6 percent, tariff cuts related to the Uruguay Round continued, and customs fees were reduced from 4 percent to 2 percent. A number of other export distortions remained, such as export subsidies for agricultural products. Finally, exchange restrictions were gradually liberalized, notwithstanding some tightening in 1990–92, and the currency became convertible in 1996. The surcharge was removed in mid-1997.

Table 18.Hungary: Restrictiveness of Trade Regime
Source: IMF staff estimates.Scale: 1 (least restrictive) to 10 (most restrictive). n.a. denotes not applicable.
Source: IMF staff estimates.Scale: 1 (least restrictive) to 10 (most restrictive). n.a. denotes not applicable.

Targets and Implementation Under Fund Programs

Two of the four Fund-supported programs contained trade reform. The 1991 EFF required Hungary to liberalize import licenses, including a shift of the CMEA trade to world prices, and to increase the quota on consumer goods imports. The 1996 Stand-By Arrangement included the phasing out of the import surcharge and a commitment to reduce the number of consumption goods subject to the import quota.

All of the above-mentioned targets were met and sometimes exceeded. The trade reform under the 1991 EFF was implemented despite the fiscal difficulties encountered in the programs, and Hungary reduced the import surcharge ahead of the schedule agreed with the Fund in 1996.

Factors Affecting Design and Implementation of Trade Reform in Fund Programs

External factors affecting trade reform usually emerged in the design stage. Balance of payments concerns may explain the lack of trade reform in the program supported by the 1990 Stand-By Arrangement. The design of the 1991 EFF was influenced positively by the improved current account position, and an acknowledged need for structural reform by the government. Trade reform was closely coordinated with the World Bank. Since 1992, regional trading arrangements may have influenced Hungary’s willingness to undertake broader nondiscriminatory trade reform, as trade policy was linked to its free trade agreement with the European Union (EU). The 1993 Stand-By Arrangement did not contain any trade reform. Fund staff had suggested that further trade reform was needed, and the lack of trade reform may have been influenced by the difficult fiscal and balance of payments positions as well as the substantial increase in unemployment in the preceding years. The design of the 1996 Stand-By Arrangement, in which Fund staff had initially proposed more substantial trade reform, may have been influenced by the fragile fiscal situation and protectionist lobbying. The agreed phase-out of the surcharge was more gradual than that proposed by the Fund staff. The desire to protect some local industries and ongoing liberalization in the GATT-WTO context were cited as reasons for not eliminating the remaining NTBs. On trade reform implementation, the government’s commitment to liberalization was an important factor, particularly in the early 1990s when Hungary implemented trade reform despite the rest of the program running into difficulties, and at other times when trade liberalization exceeded program targets. Also, as mentioned above, the phasing out of the surcharge in 1996 was implemented faster than was agreed in the program.


The most important factor promoting trade reform in Hungary may have been the government’s commitment to undertake reform. Its desire to dismantle central planning was a major policy objective and influenced liberalization, following the collapse of the CMEA trading system. Subsequently, NTB reduction has been slower and progress with tariff reform mixed. Tariffs were lowered in the context of WTO obligations, but a surcharge was introduced. Hungary’s desire to liberalize only in the context of the GATT-WTO and regional agreements may have slowed down the pace of trade reform. For example, the authorities were reluctant to undertake further reductions in the quota on consumer goods imports partly because this could have reduced leverage in international trade negotiations. Fiscal and balance of payments difficulties critically affected overall program implementation and, at times, the pace of trade reform.

Sri Lanka

During the period 1990–96, Sri Lanka had one Fund arrangement. The ESAF was approved in September 1991 and covered the period through July 1995. It continued the structural reforms initiated in the context of the 1988–91 SAF.

Sri Lanka had a moderately restrictive trade system initially, classified as a 7 on the scale of restrictiveness. The average tariff was estimated at 16 percent in 1990, and the maximum rate was 60 percent. In addition, there was a 2½ percent stamp duty on import letters of credit and a markup of 25 percent was applied to the c.i.f. value of imported (but not domestic) goods for the calculation of domestic indirect taxes. There were a number of specific duties that in some cases had ad valorem equivalents in excess of 100 percent. As a result, combined average tariff protection was probably 20–25 percent at the outset of the ESAF. NTBs were significant but not pervasive. A significant number of items were subject to import licensing and to import monopolies and several items were subject to export licensing. Export taxes applied principally to Sri Lanka’s traditional exports of tea, rubber, and coconut products.

Substantial liberalization of the trade regime was achieved in 1990–96 moving Sri Lanka to a rating of 2 on the scale of restrictiveness (Table 19). By the end of the ESAF in 1995, Sri Lanka had an average tariff estimated at 8 percent. The maximum rate was reduced to 35 percent. However, the markup of 25 percent applied to the c.i.f. value of imported (but not domestic) goods in the calculation of domestic indirect taxes probably pushed overall average tariff protection to above 10 percent. Tariff reform lost momentum after the end of the program, and a planned shift to a two-band tariff system, with a maximum rate of 25 percent, was delayed in part for fiscal reasons. Import NTBs were largely removed during the program, and in 1996, import licensing was further scaled back. By the end of the program period, export licensing was restricted to items justified for environmental or cultural reasons only. Export taxes had been largely eliminated in the course of the program.

Table 19.Sri Lanka: Restrictiveness of Trade Regime
Source: IMF staff estimates.Scale: 1 (least restrictive) to 10 (most restrictive). n.a. denotes not applicable.
Source: IMF staff estimates.Scale: 1 (least restrictive) to 10 (most restrictive). n.a. denotes not applicable.

Targets and Implementation Under the Fund Program

The ESAF sought to reduce average tariffs while reducing dispersion and the maximum rate from 50 percent to 35 percent. Also, it sought to eliminate import and export NTBs as well as export taxes. The program had extensive conditionality on trade reform to signal its importance. The first stage of Sri Lanka’s trade reform, including a reduction in the maximum rate and in the number of tariff bands, was monitored by a structural performance criterion in the first annual arrangement. The second stage of trade reform was monitored by a structural benchmark in the second annual arrangement, and the third stage, covering the elimination of most import and export licensing requirements, was a prior action for the third (and final) annual arrangement.

Trade reform was generally implemented as envisaged, aside from some minor deviations and delays. Some tariff reform measures planned for April and June 1993 were implemented in November 1993 as part of the 1994 budget. A more substantial delay was the reduction in the maximum tariff rate to 35 percent, which did not take place until early 1995. However, the trade regime at the end of the program was essentially that envisaged in the program.

Factors Affecting Design and Implementation of Trade Reform in Fund Program

Sri Lanka’s trade reform was largely implemented as initially planned, which reflected, inter alia, the commitment of the authorities. Toward the end of the ESAF, negotiations became more protracted. Trade reform, however, does not seem generally to have been a significant factor in delaying the negotiations.

Fiscal policy was probably an important factor influencing the design, negotiation, and implementation of trade reform. The program ultimately went off track for fiscal reasons in 1994. It seems unlikely that trade reform in the program, which included reductions in tariffs and export taxes, was considered in isolation given Sri Lanka’s dependence on trade taxes and its large fiscal deficits. The World Bank was involved by providing technical assistance on trade reform.

Balance of payments considerations and membership in RTAs do not appear to have affected the pace of trade reform. The reason for this may be related to Sri Lanka’s access to highly concessional external financing, including grants. Although Sri Lanka is a member of the South Asian Preferential Trading Arrangement, the preferences extended by (and to) Sri Lanka under this RTA have been limited so far.70

It is not clear whether other factors were behind the slippages in trade reform in 1994. The elections may have affected the pace of trade reform as election-related spending led to fiscal pressures and political considerations in an election year promoting protectionist interests.


Sri Lanka had begun to liberalize its trade regime in 1977. An important factor in its relative success at liberalization in the 1990s was the commitment of the successive governments (of both major parties) to trade reform. Although at times fiscal policy may have affected the pace of trade reform, the serious fiscal deficits in Sri Lanka have not stopped trade reform, including the implementation of some revenue-reducing elements of trade reform.


During the 1990–96 period, Zambia had three Fund-supported programs. The RAP of July 1992 followed the RAP of 1991, which went off track. The 1992 RAP was negotiated against the background of a severe shortfall in food supplies and of worsened fiscal and balance of payments positions. Following the accumulation of rights and the repayment of Zambia’s arrears to the Fund, the SAF-ESAF-supported program came into effect in December 1995.

Zambia started the period with a moderately restrictive trade regime that ranked 7 on the scale of restrictiveness (Table 20). In late 1990, as part of a World Bank—supported effort, the maximum tariff rate was gradually reduced to 50 percent (from 100 percent), while the minimum rate was increased to 15 percent (from zero). Tariff protection, however, was also affected by the sales tax on imports, as some imports were exempted from the sales tax while other imports were subject to a de facto higher sales tax rate. In addition, a licensing levy of 5 percent applied to imports subject to licensing. Zambia granted numerous tariff exemptions under the Investment Code as well as for imports by the government, the state mining company, and numerous (privatized) companies and charitable institutions. Finally, mining exports were taxed at 20 percent, although exemptions were allowed, and the state mining company was exempted, as it was already paying the mineral resources levy.

Table 20.Zambia: Restrictiveness of Trade Regime
Source: IMF staff estimates.Scale: 1 (least restrictive) to 10 (most restrictive). n.a. denotes not applicable.
Source: IMF staff estimates.Scale: 1 (least restrictive) to 10 (most restrictive). n.a. denotes not applicable.

Zambia implemented an impressive trade reform program in 1990–96 and currently has one of the more liberal trade regimes in Africa, ranking 3 on the scale of restrictiveness. Zambia’s participation in the CBI trade and investment initiative has been an important factor in furthering reform. The 1996 budget included further tariff reform measures significantly ahead of schedule according to Zambia’s commitments under the CBI and Common Market for Eastern and Southern Africa (COMESA). As a result, the maximum tariff rate was reduced to 25 percent and the (general) minimum rate to 5 percent in February 1996, although some inputs and machinery continue to be rated zero. There are now only four tariff bands (zero, 5, 15, and 25 percent) and the unweighted average tariff rate is estimated at 13.6 percent. The import declaration fee of 5 percent, however, was not abolished at the end of 1996 as envisaged, but the authorities have indicated their intention to remove this surcharge at the end of 1997. Zambia’s trade regime is essentially free of import and export NTBs.

Targets and Implementation Under Fund Programs and in Nonprogram Years

The Fund programs sought to move Zambia by 4 points on the scale of trade restrictiveness during the period. The 1992 RAP program was mainly concerned with liberalizing Zambia’s restrictive exchange regime that severely restricted trade. There was a clear understanding that as Zambia had made considerable progress with trade reform, recurring fiscal slippages imposed limits on the tariff reform that could be pursued, and that the World Bank would take the lead. Toward the end of the program, Zambia’s obligations to the CBI were the basis for future reforms. The only trade policy target monitored by binding conditionality was the abolition of the import licensing requirement in June 1993, which was observed on a timely basis. Fund programs also envisaged introduction of a VAT, achieved in 1995.

During the 1992 RAP, the import licensing fee equal to 10 percent was abolished in January 1993, and in June 1993 all licensing requirements were also removed. In 1994, the maximum tariff was reduced to 40 percent. In October 1995, faced with fiscal disequilibria, the government introduced a 5 percent import surcharge. Despite this reversal, in early 1996 Zambia further liberalized its trade regime to conform, ahead of the October 1998 deadline, with its CBI obligations. In general, Zambia implemented most commitments in the programs.

Factors Affecting Design and Implementation of Trade Reform in Fund Programs

The need to address fiscal problems and liberalize the exchange regime was accorded priority over trade issues in the Fund programs. Fund-Bank collaboration in trade reform was effective. There was also an implicit acceptance by the IMF staff of the authorities’ commitment to trade reform.


Zambia has succeeded in implementing an impressive trade reform program since the late 1980s. Starting with a very complex and highly restrictive exchange and trade regime, it now has one of the more liberal trade regimes in Africa. This was done in advance of the country’s obligations to the CBI and COMESA and generally exceeded the liberalization envisaged under Fund programs. Maximum tariffs were reduced and export restrictions were largely removed. The only setback was the introduction in 1995 of the 5 percent import surcharge.

The authorities’ commitment to trade reform has been demonstrated by their actions to liberalize trade ahead of commitments to the Fund and other institutions. Fund-Bank collaboration also appears to have supported trade reform. Zambia’s participation in the CBI has promoted trade reform. Recurrent fiscal pressures however, put strains on reform efforts. In 1995, the imposition of a surcharge was a response to redress fiscal slippages.

Zambia started the period with severe external and fiscal imbalances and a highly regulated domestic economy. Initially, Fund programs focused on fiscal consolidation and exchange liberalization to pave the way for the rapid and durable trade liberalization.


During the period 1990–96, Zimbabwe had a program supported by closely related Fund arrangements. The EFF and ESAF arrangements covered the period July 1992 to June 1995. It replaced the three-year EFF that had come into effect in January 1992 and was canceled when Zimbabwe became eligible to use ESAF resources. The policy content of the EFF and ESAF was similar to that of the original EFF. The program expired without the scheduled third annual arrangement.

In 1990, Zimbabwe’s economy was heavily regulated. The trade regime ranked 10 on the scale of restrictiveness. Tariffs averaged some 20 percent and were supplemented by a 20 percent import surcharge. Thus, average tariff protection, including the surcharge, was some 40 percent. Also, import licensing was used to ration foreign exchange and exporters were required to surrender their export earnings at the (overvalued) official exchange rate. The import and export of many agriculture products were restricted to marketing boards.

Zimbabwe’s trade system was somewhat liberalized in 1990–96, owing mainly to the near elimination of its import licensing system. As a result, Zimbabwe moved to a rating of 8 on the scale of trade restrictiveness (Table 21). Notwithstanding the reductions in the import surcharge, tariffs remained high and averaged some 30 percent by the end of the period.

Table 21.Zimbabwe: Restrictiveness of Trade Regime
Source: IMF staff estimates.Scale: 1 (least restrictive) to 10 (most restrictive). n.a. denotes not applicable.
Source: IMF staff estimates.Scale: 1 (least restrictive) to 10 (most restrictive). n.a. denotes not applicable.

Targets and Implementation Under Fund Programs

The programs aimed at moving Zimbabwe by 4 points to 6 on the scale of trade restrictiveness. The main trade reform measures were liberalizing Zimbabwe’s import licensing system as well as eliminating the import surcharge. This was to be achieved by expanding the unrestricted Open General Import License (OGIL) system and the Export Retention Scheme (ERS). The products on the OGIL list could be imported by anyone and the list of eligible products was expanded. The ERS allowed exporters to retain a certain percentage of their export receipts, to be used to import eligible products. ERS expansion increased the retention share and expanded the list of eligible products for importation using foreign exchange retained under the ERS. Early in the EFF and ESAF arrangements, with support from IMF staff, the authorities decided to place greater emphasis on ERS expansion as the main vehicle of import liberalization and the envisaged liberalization of import licensing was achieved. Commitments to reduce tariff protection, however, were largely limited to the elimination of the import surcharge, with a reduction from 20 percent to 10 percent by June 1993 and elimination by June 1995.

Implementation of the EFF arrangement was weakened by a number of factors, including severe drought, and therefore the envisaged trade reform was not achieved. During the successor EFF and ESAF arrangements, not all trade reform targets were met. Trade policy commitments monitored by binding conditionality were satisfactorily implemented: the ERS expansion was achieved in a more ambitious form, following the decision to concentrate on the ERS and to discontinue the OGIL. The reduction of the import surcharge to 10 percent was delayed, however, until 1994, and its elimination was not achieved.

Factors Affecting Design and Implementation of Trade Reform in Fund-Supported Programs

Zimbabwe’s relatively weak overall economic performance in the 1980s was an important factor influencing economic reform efforts, including trade reform. This could explain why no substantial differences of opinion seem to have emerged during the program design stage. The authorities had charted a course of reform starting in 1990, and IMF and World Bank involvement provided further credibility to the measures. However, the very difficult fiscal situation and the impact of the drought may have influenced the pace of trade liberalization in the design stages. Reform of the foreign exchange and import licensing regime was clearly recognized as a high priority. Tariff reform was accorded a lower priority, and there were some misgivings about tariff reform, given the difficult fiscal situation. Furthermore, with the elimination of import licensing, pressure for import protection through tariffs increased.


The authorities made a good deal of progress liberalizing imports through elimination of the import licensing scheme but were reluctant to move rapidly on trade reform, perhaps because of the impact of the drought, which affected the political economy of trade reform. The main failure in the area of trade reform was the inability to adhere to the schedule of surcharge reduction and elimination. This may have been part of a broader problem related to the overall macro-economic targets, including fiscal slippages.


World Economic and Financial Surveys

This series (ISSN 0258-7440) contains biannual, annual, and periodic studies covering monetary and financial issues of importance to the global economy. The core elements of the series are the World Economic Outlook report, usually published in May and October, and the annual report on International Capital Markets. Other studies assess international trade policy, private market and official financing for developing countries, exchange and payments systems, export credit policies, and issues discussed in the World Economic Outlook. Please consult the IMF Publications Catalog for a complete listing of currently available World Economic and Financial Surveys.

World Economic Outlook: A Survey by the Staff of the International Monetary Fund

The World Economic Outlook, published twice a year in English, French, Spanish, and Arabic, presents IMF staff economists’ analyses of global economic developments during the near and medium term. Chapters give an overview of the world economy; consider is-sues affecting industrial countries, developing countries, and economies in transition to the market; and address topics of pressing current interest.

ISSN 0256-6877.

$36.00 (academic rate: $25.00); paper.

1997 (Dec). ISBN 1-55775-714-3 (English only). Stock #WEO-1797.

1997 (Oct.). ISBN 1-55775-681-3. Stock #WEO-297.

1997 (May). ISBN 1-55775-648-1. Stock #WEO-197.

Official Financing for Developing Countries

by a staff team in the IMF’s Policy Development and Review Department led by Anthony R. Boote and Doris C. Ross

This study provides information on official financing for developing countries, with the focus on low-income countries. It updates the 1995 edition and reviews developments in direct financing by official and multilateral sources.

$25.00 (academic rate: $20.00); paper.

1998. ISBN 1-55775-702-X. Stock #WEO-1397.

1995. ISBN 1-55775-527-2. Stock #WEO-1395.

Issues in International Exchange and Payments Systems

by a staff team from the IMF’s Monetary and Exchange Affairs Department

The global trend toward liberalization in countries’ international exchange and payments systems has been widespread in both industrial and developing countries and most dramatic in Central and Eastern Europe. Countries in general have brought their exchange systems more in line with market principles and moved toward more flexible exchange rate arrangements in recent years.

$25.00 (academic rate: $20.00); paper.

1995. ISBN 1-55775-480-2. Stock #WEO-895.

Staff Studies for the World Economic Outlook

by the IMF’s Research Department

These studies, supporting analyses and scenarios of the World Economic Outlook, provide a detailed examination of theory and evidence on major issues currently affecting the global economy.

$25.00 (academic rate: $20.00); paper.

1997. ISBN 1-55775-701-1. Stock #WEO-397.

International Capital Markets: Developments, Prospects, and Key Policy Issues

by an IMF staff team led by David Folkerts-Landau with Donald J. Mathieson and Garry J. Schinasi

This report provides a comprehensive survey of recent developments and trends in the mature and emerging capital markets, including equities, bonds, foreign exchange, and derivatives, and banking systems. It focuses on the implications of European Economic and Monetary Union (EMU) for financial markets and the management of external liabilities of emerging market countries.

$25.00 (academic rate: $20.00); paper.

1997. ISBN 1-55775-686-4. Stock #WEO-697.

Private Market Financing for Developing Countries

by a staff team from the IMF’s Policy Development and Review Department led by Steven Dunaway

The latest study surveys recent trends in flows to developing countries through banking and securities markets. It also analyzes the institutional and regulatory framework for developing country finance; institutional investor behavior and pricing of developing country stocks; and progress in commercial bank debt restructuring in low-income countries.

$25.00 (academic rate: $20.00); paper.

1995. ISBN 1-55775-526-4. Stock #WEO-1595.

Toward a Framework for Financial Stability

by a staff team led by David Folkerts-Landau and Carl-Johan Lindgren

This study outlines the broad principles and characteristics of stable and sound financial systems, to facilitate IMF surveillance over banking sector issues of macroeconomic significance and to con-tribute to the general international effort to reduce the likelihood and diminish the intensity of future financial sector crises.

$25.00 (academic rate: $20.00); paper.

1998. ISBN 1-55775-706-2. Stock #WEO-016.

Trade Liberalization in IMF-Supported Programs

by a staff team led by Robert Sharer

This study assesses trade liberalization in programs supported by the IMF by reviewing multiyear arrangements in the 1990s and six detailed case studies. It also discusses the main economic factors affecting trade policy targets.

$25.00 (academic rate: $20.00); paper.

1998. ISBN 1-55775-707-0. Stock #WEO-1897.

Available by series subscription or single title (including back issues); academic rate available only to full-time university faculty and students. For earlier ediitons please inquire about prices.

The IMF Catalog of Publications is available on-line at the Internet address listed below.

Please send orders and inquiries to:

International Monetary Fund, Publication Services, 700 19th Street, N.W.

Washington, D.C. 20431, U.S.A.

Tel.: (202) 623-7430 Telefax: (202) 623-7201



Article I (ii) states that the Fund is “To facilitate the expansion and balanced growth of international trade, and to contribute thereby to the promotion and maintenance of high levels of employment and real income and to the development of the productive resources of all members as primary objectives of economic policy.”

The Fund and the WTO have concluded a Cooperation Agreement that came into effect on December 9, 1996.

Exchange rate policy and exchange reform are linked in important ways to trade policy but generally fall outside the scope of trade liberalization, narrowly defined, and hence are generally excluded from this study, except where such restrictions act as binding trade restrictions.

Nontariff barriers are defined here as measures that restrict trade by either limiting quantities that may be traded or inserting administrative discretion in the export-import process that acts as a barrier to trade. Such NTBs include quantitative restrictions, quotas, bans, restrictive licensing requirements, restrictive foreign exchange practices, state trading monopolies, restrictive technical and phytosanitary standards, and restrictive customs procedures.

Uniform tariffs are preferable on political economy grounds since they reduce the incentive to lobby for a higher tariff rate. However, uniform nominal tariff rates need not imply uniform effective rates of protection if, for instance, domestic taxes are higher in some sectors; see Thomas, Nash, and others (1991).

This includes equal tax treatment of domestic and foreign entities, equal tax treatment of imported and domestic goods and services, and equal technical and phytosanitary standards.

Often import quantitative restrictions are replaced by tariffs.

Political economy forces may be working in favor of trade liberalization, for example, to foster closer relationships with neighboring countries.

This assessment of trade policy does not imply any judgment of overall program content. Throughout this study, the terms “Fund-supported programs” and “Fund programs” are used interchangeably.

The focus on multiyear arrangements introduces some bias in the sample group of arrangements toward Structural Adjustment Facility (SAF) and Enhanced Structural Adjustment Facility (ESAF) programs with low-income countries. These account for 19 of the 30 arrangements reviewed.

There were two arrangements for Burkina Faso: a SAF in 1991 and an ESAF in 1993. The policy programs supported by these arrangements were distinct and separate. There were three arrangements for Zimbabwe but were in support of one policy program.

Thus it does not cover significant trade reform efforts that may have occurred preceding or following the program under review, as was the case, for example, in Guyana, Jordan, Panama, and Peru.

Appendix I discusses the classification of tariff and NTB regimes, how the 10-point scale combines tariff and NTB measures, and how other studies have quantified the overall restrictiveness of trade regimes in developing countries.

Other NTBs include safeguard actions, antidumping and countervailing duties, reference prices, quality controls, and indicative export prices.

It should be noted that analysis of the initial conditions of the trade system refers to the state of the trade policy regime before implementation of any prior actions under the program. Prior actions are included in the set of measures considered in assessing ambitiousness of program targets.

For the analysis, countries with index classifications of 8–10) can be considered restrictive; classifications of 5–7, moderate; and classifications of 1–4, open.

The finding that trade regimes were restrictive at the outset of programs is consistent with evidence in the IMF’s 1997 ESAF review (Bredenkamp and others, 1998) that, starting from a situation of relatively restrictive regimes, countries using ESAF resources made only moderate progress in trade liberalization between 1985 and 1990; it should be noted that the coverage of countries differed in the two studies.

In assessing the program objective for trade liberalization, all measures targeted during the program were taken into account, including prior actions as well as measures specified after the arrangement went into effect: for example, as part of the second- or third-year programs.

In principle, the index could understate the degree of liberalization targeted, for example, in the most restrictive trade regimes, when changes are introduced, which might be considered an important first step toward longer-term liberalization but are insufficient to move the country below the highly restrictive category. In practice, however, this was rarely the case.

Some instances of temporary increases in protection during the program period were eliminated by the end of the program.

Prior actions, performance criteria, benchmarks, and reviews.

Examples include medium-term targets that sought to “rationalize the tariff structure” or “reduce the number of products subject to restrictive licensing” without specifying the extent or details of reform envisaged or the implementation dates.

These countries are illustrative of highly successful open economies and are not necessarily typical of all successful trade-oriented economies.

The issue of slippages in trade policy reforms over the medium term is not addressed in the review. However, the case studies investigated trade reform efforts over a seven-year period and found only limited indication of slippages.

The review found no obvious common thread in these programs that could explain the lack of movement in trade restrictiveness.

Under WTO rules, member countries have recourse to balance of payments safeguards, that is, provisions permitting member countries to impose temporary restrictions on trade in goods and services for balance of payments purposes.

The analysis of this section is based on reviews of the literature, on Fund and Bank staff experiences, particularly Fund staff’s reviews of tax policy and administration, and on the review of Fund arrangements included in this study.

If the average tariff is very high, however, the impact of a reduction may be ambiguous.

Rajapatirana (1996) assessed the impact of trade policies in Latin America in the 1970s and 1980s and found three cases in which trade liberalization, largely through liberalization of quantitative restrictions, resulted in an increase in revenue from trade taxes (Chile, Colombia, and Costa Rica).

Nogués and Gulati (1992, p. 70) found that in Chile, reductions in very high tariffs were associated with increases in tariff revenue and found “no evidence that higher tariff protection is associated with higher tariff and tax revenues” in a sample of five Latin American countries.

Throughout the study, it is assumed that trade reform should not affect the overall fiscal policy stance, which is set to achieve savings and investment objectives consistent with targeted growth and inflation.

From a trade perspective, a VAT may be preferable to a cascading sales tax since the latter tends to discriminate in favor of imports and against exports.

Dixit (1985) provides a detailed explanation of the application of optimal tax theory to taxes on international trade.

This is an important source of efficiency gains, because economic distortions tend to increase more than proportionately with increases in tax rates.

Fiscal conditions are considered as a possible constraint on trade reform if trade taxes were a high proportion of total taxes (over 30 percent), tax revenue was low (less than 10 percent of GDP), the targeted reduction in the budget deficit over the period of the arrangement was high (over 4 percent of GDP), or if there was no broad-based consumption tax such as a VAT. These cutoff levels were based on a review of data for developing countries and Fund programs.

The ESAF for Bangladesh was followed by a program monitored by the IMF staff under enhanced surveillance.

IMF staff have usually argued that since WTO accession negotiations involve bindings of maximum rather than applied rates, the WTO accession process should not constrain more ambitious reforms that are economically desirable.

While not cited in the review, Zambia’s participation in the Cross-Border Initiative (CBI) has influenced its trade policy, as noted in the case studies below.

For more details on the selection of country case studies, see Appendix I.

The case studies for each of the six countries are summarized in Appendix II.

These prior actions are described in Appendix II.43 The amount of trade liberalization targeted at this stage is difficult to measure on the index of restrictiveness. Therefore, broad assessment is made on trade liberalization proposed in terms of a general reduction in the scope of restrictiveness.

The amount of trade liberalization targeted at this stage is difficult to measure on the index of restrictiveness. Therefore, broad assessment is made on trade liberalization proposed in terms of a general reduction in the scope of restrictiveness.

Even in the sixth country, Sri Lanka, fiscal considerations were cited by the authorities as affecting the pace of implementing trade reform after the ESAF arrangement, although these have not stopped trade reform.

A similar point is made in a recent study of liberalization in Latin America. Primo Braga, Nogués, and Rajapatirana (1997, p. KM) note the substantial liberalization in the region since the mid-1980s but also observe some slowdown or even reversal: “Many countries that faced fiscal deficits recently have raised tariffs. However, it is not at all clear that fiscal adjustment in the form of higher tariffs will help these countries meet revenue targets.”

However, this need not imply that slower reforms were not justified, since other constraints such as political economy factors may have required slower liberalization.

The indicators used were the existence of a VAT, capacity for revenue collection (tax revenue/GDP), dependence on trade taxes for revenue (trade taxes/tax revenue), the targeted reduction in the fiscal deficit/GDP, and the extent of exemptions.

The surcharge, however, was subsequently removed somewhat ahead of the program schedule.

For example, while the average tariff was 89 percent, actual collection of import duties was only 16 percent of imports in 1990. The ratio may be misleading due to the existence of donor-financed (duty-free) imports or an export processing zone, membership in an RTA, and various other factors. Nevertheless, reduction of exemptions to import and sales taxes was stated only as a medium-term target, without any specificity.

The only binding conditionality on trade reform in the initial program was a prior action related to the tariffication of NTBs covering 5 percent of tariff lines, leaving another 20 percent still subject to NTBs.

Collection of import duties was about 9 percent of imports while the average tariff was estimated to be about 35 percent in 1990/91.

As a general rule, trade restrictions are not the appropriate policy option for addressing external imbalances, because they add to economic distortions and do not address the cause of the imbalance. This principle underlies the legal foundation of the WTO as well as the Fund’s Articles of Agreement. External imbalances should be addressed by appropriate macroeconomic and structural policies that deal with the underlying investment-saving imbalances. The use of trade restrictions can be justified only in exceptional circumstances, and then only as a temporary short-term measure.

The Uruguay Round reduced only one component of Hungary’s tariff structure and did not offset the new import surcharge that was introduced in 1995. When the Uruguay Round cuts started to be implemented, the average tariff actually increased in Hungary from 18 percent to 22 percent. The Uruguay Round NTB cuts did not change Hungary’s NTB import coverage ratio.

Credit should be given to the CBI for trade reform in Zimbabwe, although the Fund-supported programs in the case study preceded the start of the CBI.

The CBI also resulted in trade-enhancing reforms in areas such as investment promotion, reduced procedural impediments, and general efforts to curtail impediments to cross-border trade and investment.

New border taxes cannot be introduced under WTO rules, while existing tariffs can be increased only within existing bindings.

The ESAF arrangements with Burundi and Rwanda, both of which began in 1991, are excluded.

The preferred approach was to use, where possible, an unweighted average based on statutory tariff rates and including any other duties and charges applicable to imports. An average of statutory tariff rates is preferable to an average based on customs duties collected since the latter reflect (often extensive) exemptions. An unweighted average is preferable to a trade-weighted average since items with high tariffs would likely have small trade weights. Other duties and charges should be included because they have the same restrictive effects as tariffs.

The ad valorem equivalent of an import quota is the rate of ad valorem tariff that would yield the same import quantity as the quota. There are many circumstances in which import quotas and import tariffs are not equivalent, including where market structures are imperfectly competitive or where trading partners may retaliate against increases in tariffs or quotas; see Bhagwati and Srinivasan (1983) for further discussion and references. Notwithstanding these well-known theoretical difficulties, price equivalents of NTBs are often computed by comparing domestic with world market prices. Quantification of the price equivalent of an export restriction is often accomplished in a similar manner.

Trade or production coverage of NTBs measures the presence of NTBs but does not fully capture their restrictiveness.

NTBs include quantitative restrictions, state trade monopolies, restrictive foreign exchange practices that affect the trade regime (e.g., a surrender requirement at a nonmarket exchange rate), quality controls and customs procedures that act as trade restrictions. This definition of NTBs is broader than used in the ESAF review (Bredenkamp and others, 1998) and in an earlier Fund study (Kirmani and others, 1994b).

Information used to construct these indices often included nominal and effective rates of protection, the prevalence and degree of restrictiveness of QRs, the degree of antiexport bias, the extent of divergence of the exchange rate from its equilibrium level, the structure of imports (especially the share of final consumer goods in total imports), subjective evaluations of the degree of friction in the administrative machinery, and so on.

A country was judged to have had extreme distortions resulting from an export marketing board based on results from a World Bank study, Husain and Faruquee (1994), which covers only African countries.

Additional trade liberalization has been implemented so far in 1997 with the maximum tariff rate (and other intermediate rates) reduced by 5 percentage points. Further liberalization is planned for 1998.

There are import bans on textiles and garments estimated to cover less than 5 percent of tradable goods output. As part of the Uruguay Round agreement, Egypt has agreed to eliminate the import ban on textiles by January 1, 1998 and on clothing by January 1, 2001. An import ban on poultry was eliminated in July 1997.

The increase of the minimum tariff would have breached some of Egypt’s GATT bindings, and therefore not all of this increase took place.

The maximum tariff rate and tariff rates above 50 percent were reduced by 15 percentage points, and tariff rates of 40–50 percent by 10 percentage points.

The first set of tariff reductions was implemented on July 1, 1997.

The higher rate surcharge was reduced from 4 percent to 3 percent on July 1, 1997.

Sri Lanka extends concessions on 31 tariff lines only and the margin of preference ranges between 10 percent and 20 percent of the general tariff rate. Concessions to Sri Lanka by other members are similarly minor.

    Other Resources Citing This Publication